Membrane proteins represent one of the greatest structural genomics challenges today. These proteins are of unquestionable importance as they exist at the interface between a cell's exterior and interior. They represent the majority of today's drug targets. And yet we have so little structural information on these proteins. Here we present a plan that will enhance the rate at which membrane protein structures will he achieved by an order of magnitude and likewise reduce the cost per structure by an order of magnitude. 13 collaborating investigators from 8 institutions bring a wealth of expertise in membrane protein production and a diversity of technologies to bear on this important problem. In bringing such a team together great opportunities exist for synergy. Knowledge will be spilled across disciplinary boundaries facilitated by a substantial effort to make communication both easy and productive. Our structural efforts will be targeted against the membrane proteins of Mycobacterium tuberculosis, the world's leading cause of infectious mortality. An effort will be made to express each of the nearly 1200 Open Reading Frames that represent membrane proteins. It is anticipated that 10 to 20 percent of these proteins will be expressed in first two years of funding. For each expressed protein, sample preparation will be attempted for at least two different technologies, such as 3D crystals for X-ray crystallography and aligned bilayers for solid state NMR spectroscopy. It is our stated aim that this team will solve 50 membrane protein structures. This would nearly triple the number of membrane protein structures in the Protein Data Bank. More importantly, it would be a major step forward in characterizing membrane protein folds and would represent a treasure trove of structural information for combating this deadly organism.